I've been considering this question, and found many people asking the same (or something similar) online, but none of the answers seemed to address the core point or at least I wasn't able to make sense of them in that regard. I'm looking for a layman-friendly explanation.

I'll copy the clearest exposition of this question I found, from here:

What is it that causes an electromagnetic wave to oscillate? I.e. what is the medium that constrains it and pulls it back from one side to the other?

Discounting pressure waves such as sound, as they push backwards and forwards through air, other waves, with amplitude, need something to constrain them. For instance waves in water are constrained by surface tension and gravity so at the end of the "up" gravity and surface tension pulling the water flat pull it back down and vice versa at the end of the "down".

So what is doing that with visible light, gamma rays, microwaves etc? What pulls their energy back and forth causing them to oscillate?

One of the answers there claimed "The electric field constrains the magnetic field and vice versa." Is this true? I didn't see anyone else making a similar claim.

The above pretty much sums up my question, but for completeness' sake, I'm also including below references to some related threads I found:

This guy asked "Do photons oscillate?". The responses explained nicely how his conception of photons was incorrect, but didn't clarify his question of why the wave forms.

Somebody asked a pretty similar question on reddit: "Why do photons oscillate?", adding "I don't understand what makes them swing back and forth." -- again, the answer explained that photons aren't really "particles of light", and don't oscillate, but again no satisfactory answer to the "why" question was included.

Finally, this page has a great applet demonstrating the basic principle of how disturbances in one charge propagate to nearby ones. The oscillation is provided by a spring for illustration purposes. The text naturally raises the question: "Yeah, that all makes sense, but don't expect me to believe that particles as small as electrons are attached to springs. How is [the first] electron made to wiggle -- I mean, how is its speed or direction of motion changed?" -- but I couldn't find the answer in the following pages.

EDIT: I emphasized some passages above to make it clear I am not assuming the EM field moves in space (I realize what is changing is the amplitude over time), and that what I'm asking is indeed, to quote Bjorn Wesen's answer, "why does the source amplitude oscillate in the first place".

  • $\begingroup$ In this link motls.blogspot.com/2011/11/… the emergence of classical electromagnetic fields from zillions of photons is described mathematically. $\endgroup$ – anna v Nov 8 '15 at 11:54
  • $\begingroup$ "Do photons oscillate?" You claim that this concept was nicely explained to be incorrect. I don't think so. The only proof is someone saying that it can't be.... that's not proof. They also claim without proof that photons are not particles. On the other hand no one can physically explain what a light wave is. No one will even try because it can't be done without incorporating oscillating photons. $\endgroup$ – Bill Alsept Dec 23 '16 at 4:41

Electromagnetic waves are produced by changing (time variable) electric and/or magnetic fields, it is wrong to picture electric field extending in one direction then a particle or something else pulling it back and forcing it to move in another direction.

This example is more appropriate, take radio waves, these are emitted by rapid acceleration of charges via a strong alternating electric current in an antenna/conductor. The rapid acceleration of electrons i.e. alternating current induces a time dependent magnetic field at some distance, a changing magnetic field has a virtue of inducing an electric field, since the magnitude of magnetic field changes with time so does the value of electric field, the process repeats and wave propagates forward by producing alternating fields one after the other.

This self sufficiency of electromagnetic waves also enables them to propagate in vacuum in absence of any palpable medium. It would further be easier to perceive the propagation if you do not picture electron or other particle moving in any medium as they don't actually, just imagine a variation in time and not in space.

Explanation as required by comments : When I say the fields generate each other one after the other I do not mean that one field collapses and forms the other ! I was referring to the continuum of field generation one after the other, and that each field is a cause for thr other.

  • $\begingroup$ Where did I say they collapse i generating one another ? My apologies if my answer conveys that. And what do you mean by "completely wrong" ? Isn't the propogation of EM waves made possible by alternating electric and magnetic fields ? $\endgroup$ – Rijul Gupta Jan 16 '14 at 7:05
  • $\begingroup$ I think this is qualitatively correct, as the magnetic field grows so does the perpendicular to it electric field go. you might use this illustration from hyperphysics.phy-astr.gsu.edu/hbase/waves/emwavecon.html $\endgroup$ – anna v Jan 16 '14 at 7:07
  • $\begingroup$ @ Brandon : My answer and myself do not imply either of the generation of fields in or out of phase, the question requires a layman explanation, I am just telling the op that one field produces another, you can not disprove this fact as atleast at starting of propagation this happens and therefore must happen throughout propagation, also I do not say the previous field collapses. $\endgroup$ – Rijul Gupta Jan 16 '14 at 7:16
  • $\begingroup$ @ Brandon : is your downvote still valid ? Please tell me how to improve my answer to provide more accurate explanation. $\endgroup$ – Rijul Gupta Jan 16 '14 at 7:18
  • $\begingroup$ @BrandonEnright look at the equations in my link. the solutions are plane waves with changing electric and magnetic fields in the perpendicular direction. They cannot be out of phase as far as maxwell's equations go You are confusing the circular polarization which can be split up into two plane waves fulfilling maxwell's equations the two waves are out of phase with each other van.physics.illinois.edu/qa/listing.php?id=22370 $\endgroup$ – anna v Jan 16 '14 at 7:28

This is an excellent question! I think the overall confusion arises from descriptions of the EM phenomenon at a multitude of levels of understanding at the same time, each with supporting semantics.

In low-level quantum field theory, there are correlations (in 4D) in the photon field, between electrically charged fermions (like electrons and quarks), and these are always sent out and received by something. I'm hesitant to use the words "virtual photons" but this is where that terminology comes from.

When the distance between the source and receiver is large, due to interference, the only surviving contribution of these (when combined from the right source, see below) looks like the distribution of a classic "photon" with a wavelength/frequency at the speed of light and all that. If you look very close instead, you will see a lot of other strange field correlations between charged fermions that definitely don't look nice and oscillating (and don't propagate at the speed of light either).

So, you could say that the "E/B fields that intermix and create each other", or the springy analogies, are useful ways to interpret math that really arises from events that are fundamentally understood to work at a more detailed level.

To clarify another part of your question - "what oscillates" - the answer is that the quantum amplitude for sending out the field correlations that build up the "photon" oscillates at the source, and this affects the destination. There is nothing "in between" that oscillates.

It is fair to ask "so why does the source amplitude oscillate in the first place", but this has many answers, and might really require a more specific question. Roughly, think about fermion (electron) field configurations that themselves are of an oscillatory nature.

Be careful when mixing the levels of view. The various parts of the physics you are looking at have to be described in a compatible way if you are looking for intuition.

  • $\begingroup$ I'm aware that macroscopic descriptions of many physical phenomena are merely simplified models that help us reason about the underlying mechanisms at work, but not necessarily accurate depictions of what's truly happening. I'm not trying to obtain the "true" answer, but rather a useful analogy to understand the fundamental reason for the oscillations to happen at the source. In that regard, could you clarify/expand the passage "think about fermion (electron) field configurations that themselves are of an oscillatory nature"? $\endgroup$ – waldyrious Jan 20 '14 at 16:29
  • 1
    $\begingroup$ Right - well for that you don't actually need to invoke any very low-level descriptions, you just need to consider the concept of energy and the de Broglie relation between energy and frequency. An emitted physical photon oscillates at a certain frequency (has an energy) related to the loss of energy at the source. A simplified description would say that a certain bound electron in an atom itself oscillates at frequency f1, falls to f2, and emits a photon of f=f1-f2, as an example. The frequency of oscillation in essence is the definition of energy. $\endgroup$ – BjornW Jan 21 '14 at 17:42
  • $\begingroup$ Hm, thanks, that sounds like a good lead. I'll read a bit about that. Are there any (newbie-friendly) resources you'd recomend? $\endgroup$ – waldyrious Jan 21 '14 at 20:09
  • $\begingroup$ Actually, does it necessarily oscillate? I've always thought that as sinusoid waves are solutions to Maxwell equation in vacuum, by Fourier one could construct any kind of wave, including wave packets whose field was a non oscillating "bump". I thought sinusoid waves we always see were just illustrations and they would occur, as a special case, only if the charge was actually oscillating (and not just accelerating, in case of general waves). That would be the answer I'd give to this question. Could you explain me why not? $\endgroup$ – galmeida Mar 9 '14 at 23:44

In general terms as per my understanding. an EM wave is generated as follows

imagine a piece of wire of points A to B placed like this A-----B when you pass an alternating current (voltage +v 0 -v)

Emission of Electric wave.

E1: lets say for negative voltage's peak value the electron concentrates on one side of wire, lets say it at the end of wire i.e at B due to charge uneven distribution, there exist a maximum electric field emerging from A to B, lets indicate this max field to be a + peak of electric wave.

E2: lets say for Positive voltage's peak value the electron concentrates on other side of wire, lets say it at the end of wire i.e at A due to charge uneven distribution, there exist a maximum electric field emerging from B to A, lets indicate this max field to be a - peak of electric wave.

E3: when voltage is at 0 there exists no electric field. the electron may not be specifically concentrated any where. so there exits no polarity thus the field is 0.

Emission of Magnetic waves. (assuming that: u know how a magnetic field effect is observed when an electron is moving if not, just google it. its a vast topic to cover here.)

H1: lets say for negative voltage's peak value the electron concentrates on one side of wire, lets say it at the end of wire i.e at B i.e the electron moment is just about to start to other side now i.e electron velocity is min at this point so there exits min magnetic field and max E1 (as stated above)

H2: lets say for Positive voltage's peak value the electron concentrates on other side of wire, lets say it at the end of wire i.e at A i.e here also the electron moment is just about to start to other end, here also the electron velocity is min at this point so there exits min magnetic field and max E2 (as stated above )

H3: lets say for voltage in between, the electron in the wire has considerable velocity in it i.e has considerable amount of max current to flow. thus due to this movement of electron, there establishes a magnetic field maximum . i.e here the the field also exists at lower velocity of electron but gets maximum as current is max. here the electric field is 0.

here the generated magnetic field is like your left hand fingers . i.e if you place the thumb in electron moment direction i.e if electron travels from B to A then the fingers turn in anti clockwise direction and its the direction of field around the wire lets say this direction as + peak value of field and for other moment i.e from A to B you get clock wise direction lets say it -ve peak of H field

this certain point of max field , in general occurs at phase exactly 90 degrees to that of Electric field i.e H field variation is in phase with current phase and that of electric field is in phase of voltage phase ..

in space this cutting of electric and magnetic fields is also 90 degrees to each other i.e like L shape '_' -> electric field '|' -> magnetic filed i.e magnetic field cuts electric field exactly perpendicular to it ..

when this whole picture is kept in time phase and space quadrature for given A.C input you can get the picture of oscillations of the fields. hope this helps you :)

  • $\begingroup$ Stationary electrons still produce a electric fields, so E3 doesn't seem to be correct. $\endgroup$ – Kyle Kanos Nov 8 '15 at 11:43
  • $\begingroup$ its specific to situation, in general case most of them will distributed and the overall charge will be neutral. if concentrated it may produce static electricity and thats a whole new story. (just imagine how comb gets static electricity when rubbed over the silk cloth, though it presents for a while the charge distributes uniformly later resulting net zero charge thus no net field) $\endgroup$ – Srinadh kch Nov 9 '15 at 9:42
  • $\begingroup$ The problem with this explanation is that the electric and magnetic fields are out of phase. In the radiating fields, the E and B peaks are in phase. $\endgroup$ – JEB Jul 14 '18 at 1:26

At absolute zero there is no K.E. of particles in a body but at temperature > 0 K the picture changes as particles start vibrating about their positions . As we know that there are charged particles all around us , they start oscillating and oscillating charges produce Magnetic fields and these Magnetic fields are sine functions so they are also changing , which leads to oscillating Electric fields also in sine wave function . Hence an E.M.W. is formed .


Great question! Here is the simple explanation as to why electromagnetic waves oscillate.

As you probably already know, electromagnetic waves are generated by moving electrons. An electron generates an electric field. A moving electron generates a magnetic field. These combined electrical and magnetic waves reinforce one another, giving rise to an electromagnetic wave.

What causes the electromagnetic waves to oscillate? The answer is simple. It is due to the fact that matter itself is made up of atoms that are in constant motion within each matter. All matter consists of atoms in motion and these atoms, in turn, consist of positively charged protons surrounded by a cloud of negatively charged electrons. The vibrating motion of the atoms causes the cloud of electrons to oscillate and this oscillation generates electromagnetic radiation.

Since all matter contains electrons and all these electrons are in motion, all matter generates electromagnetic waves.

Since all electromagnetic radiation travels at the same velocity, the frequency and wavelength of the generated radiation depends on the frequency of the oscillating electron cloud. Thus, on average, cool objects (say those at room temperature) generate long wavelength (low frequency) radiation, while hot objects (such as the sun) generate short wavelength (high frequency) radiation.

I hope this helps.

  • $\begingroup$ You don't need atoms to generate electromagnetic radiation. As you first commented you only need electrons. The oscillating electron causes the electromagnetic field but does not explain the oscillation of electromagnetic radiation. This can be explained with oscillating photons and if their coherent they will resemble a wave. $\endgroup$ – Bill Alsept Dec 23 '16 at 7:21

My understanding of the wave effect is that the wave is only a representation of a 360° field, consider the charge traveling on a three dimensional graph. The charged wave rotates around the direction of travel. That is to say. The wave rotates in two axis as the direction of overall travel is the third axis. The representation of the a sign wave is the view of the particle as it moves towards or away from you. So a fast climb and slower curvature as it reaches and then passes it's apex, then a fast decline.

Now as to the mechanism of spin that was imparted to the force carrying particle, The best explanation that I am aware of, is that as you apply energy to a particle, you excite it's constituent components. The charge of these elemental components, as they are increased, move faster. The interaction of the separate components starts an interaction that is imparted to the entity as a whole, giving the entire particle a direction of spin. So, the particle moves away from the energizing source and according to the conservation of angular momentum, the imparted initial spin follows that direction of travel around a central axis, directly proportional to the amount of energy applied to the particle by the energizing source. That is to say, The amount of charge is directly related to the distance of spin around the central axis, as it moves in an overall direction of travel.

Hope this was clear, And again. this is my understanding of the entire process, if someone can explain it better, I am Grateful for any input.

  • $\begingroup$ Oh, and let me include, This process can be guided, such as along a wire or in a contained magnetic field, but the wavelengths can not be restricted, that is to say, that you can not shorten the wave length and still have the same frequency. This is what causes magnetic flux, the crossing of paths of different wave lengths, that can cause interference of a constructive or a destructive nature. $\endgroup$ – Dread_XO Jan 10 '18 at 0:01

First of all When you are thinking light as a classical wave then you can't think it as a photon(particle).light has dual character both wave and particle , and which character will come to us it's depend on our experiment that we will do. Now when we think light as a wave , then it is special type of wave called electro magnetic wave which is far different from mechanical wave( in which a physical particle is oscillating). In this case the electric field is varying sinusoidally and because of this variation of electric field, a sinusoidally varying magnetic field created( according to Maxwell law or more preciously Faraday law electromagnetic induction). So finally this E-M wave is just the space time SINUSOIDALLY varying electric and magnetic fields only, don't confused with classical S.H.O. Thank you


charges oscillating on a conductor generates EM wave. EM wave can also be generated when charged particles falls from states of high electric potential into states of lower electric potential much like a stone falls. When you detect the wave, what you catch are photons.

  • 2
    $\begingroup$ I agree a light wave cannot be explained without billions of coherent photons but you have not answered what is oscillating. $\endgroup$ – Bill Alsept Jan 24 '17 at 18:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.